| 309 | 0 | 463 |
| 下载次数 | 被引频次 | 阅读次数 |
随着人口老龄化进程的持续加快，我国失能老人数量不断增加，社会对长期护理保障的需求不断增长。健康状态转移概率的精确估计是长期护理需求预测和保险定价的基础。基于生命历程理论，本文将健康状态转移划分为从健康到轻度失能再到重度失能，以及从健康、轻度失能、重度失能分别到死亡的多状态转移。在多状态转移概率估计中考虑个体失能和死亡发生时间的复杂相依删失结构，构建健康、轻度失能、重度失能和从不同初始健康状态到死亡的次序多状态半参数模型，得到各阶段失能和死亡发生时间的边际分布和关联参数的一致估计，并开发不同健康状态转移概率的计算方法。基于中国老年健康影响因素跟踪调查数据，本文测算了分年龄、分性别老年人个体的各阶段失能和死亡的累积发生概率，以及健康、轻度失能、重度失能和死亡之间的状态转移概率，这一研究为预测我国失能人口规模和长期护理需求提供数据支持，也可以为长期护理保险的精准定价提供参考。
Abstract：China's aging population has led to an increase in the number of disabled older people and thus in the demand for long-term care insurance. Accurately estimating the transition probabilities among health states is fundamental to the forecasting of long-term care needs and the pricing of long-term care insurance. Grounded in life course theory, this study classifies health state transitions into multiple paths: from healthy to mild disability, to severe disability, and from each of these states to death. In estimating the multi-state transition probabilities, it is necessary to account for the complex dependency structure between time to disability and to mortality. In order to do this, a sequential multi-state semiparametric model for health, mild and severe disability, and from each of them to death is constructed. By integrating covariates into the marginal distributions, the model estimates the time to varying degrees of disability and death for the elderly, and calculates the probabilities of different health state transition paths. Based on the Chinese Longitudinal Healthy Longevity Survey, the study calculates cumulative incidence probabilities of disability and mortality for elderly individuals, stratified by age and gender, as well as transition probabilities among healthy, mild disability, severe disability, and deceased states. This research provides data support for projecting the scale of China's disabled population and long-term care demand and offers reference points for the precise pricing of long-term care insurance.
[1]陈秉正,范宸.中国老年人健康状态转移概率估计——基于两个数据库的比较分析[J].系统工程理论与实践, 2020, 40(11):2848–2860.
[2]陈鹤,赵姗姗,崔斌.长期护理保险试点财务赤字风险的评估研究——基于第一批15个试点方案的分析[J].中国卫生政策研究, 2021,14(12):42–50.
[3]陈璐,时晓爽.中国长期护理保险基金需求规模预测[J].中国人口科学, 2021(6):54–67, 127.
[4]董明英,王晓军.中国老龄人口健康受损进展与持续时间研究[J].保险研究, 2021(7):45–59.
[5]高瑗,原新.中国老年人口健康转移与医疗支出[J].人口研究, 2020, 44(2):60–72.
[6]胡宏伟,李延宇.中国农村失能老年人照护需求与成本压力研究[J].中国人口科学, 2021(3):98–111, 128.
[7]黄枫,吴纯杰.基于转移概率模型的老年人长期护理需求预测分析[J].经济研究, 2012, 47(S2):119–130.
[8]李云龙,王晓军.夫妻联合长期护理保险的定价模型与应用[J].保险研究, 2021(2):52–63.
[9]廖少宏,王广州.中国老年人口失能状况与变动趋势[J].中国人口科学, 2021(1):38–49, 126–127.
[10]刘乐平,唐爽,程瑞华.考虑状态停留时长的我国中老年人口状态转移概率测算[J].保险研究, 2020(2):102–113.
[11]仇春涓,李银环,谭昕玥,等. Markov模型框架下的重大疾病保险定价研究——基于死亡效力和发病强度的估计[J].统计研究, 2023,40(5):152–160.
[12]世界卫生组织.关于老龄化与健康的全球报告[M].世界卫生组织, 2016.
[13]宋靓珺,杨玲.老年人口健康寿命的演变轨迹及其影响因素——一项基于CLHLS的实证研究[J].人口与经济, 2020(3):57–74.
[14]张琳,汤薇.基于非齐次Markov模型的长期护理保险定价研究[J].保险研究, 2020(7):108–121.
[15]张文娟,付敏.长期护理保险制度中老年人的失能风险和照料时间——基于Barthel指数的分析[J].保险研究, 2020(5):80–93.
[16]张园,王伟.失能老年人口规模及其照护时间需求预测[J].人口研究, 2021, 45(6):110–125.
[17]朱铭来,何敏,马智苏.长期护理保险的模式选择与体系构建研究[J].中国人口科学, 2023(1):3–20.
[18]Charlotte L C, Dorina C, Anne M, et al. Operationalization of Intrinsic Capacity in Older People and Its Association With Subsequent Disability,Hospital Admission and Mortality:Results From The English Longitudinal Study of Ageing[J]. The Journals of Gerontology:Series A, 2023,78(4):698–703.
[19]Czado C, Keilegom I V. Dependent Censoring based on Parametric Copulas[J]. Biometrika, 2023, 110(3):721–738.
[20]Deresa N W, Keilegom I V, Antonio K. Copula Based Inference for Bivariate Survival Data with Left Truncation and Dependent Censoring[J],Insurance:Mathematics and Economics, 2022, 107:1–21.
[21]Deresa N W, Keilegom I V. Copula Based Cox Proportional Hazards Models for Dependent Censoring[J]. Journal of the American Statistical Association, 2023, 33(2):473–495.
[22]Ding Y, Nan B. A Sieve M-theorem for Bundled Parameters in Semiparametric Models, with Application to the Efficient Estimation in a Linear Model for Censored Data[J]. Annals of Statistics, 2011, 39(1):2795–3443.
[23]Hanewald K, Li H, Shao A W. Modelling Multi-state Health Transitions in China:a Generalised Linear Model with Time Trends[J]. Annals of Actuarial Science, 2019, 13(1):145–165.
[24]Lv Y, Yuan J, Mao C, et al. Association of Body Mass Index with Disability in Activities of Daily Living Among Chinese Adults 80 Years of Age or Older[J]. JAMA Netw Open, 2018, 1(5):1–13.
[25]Putter H, Houweilingen H C V. Frailties in Multi-state Models:Are the Identifiable?[J]. Statistical Methods in Medical Research, 2015, 24(6):675–692.
[26]Rivest L P, Wells M T. A Martingale Approach to the Copula-graphic Estimator for the Survival Function under Dependent Censoring[J]. Journal of Multivariate Analysis, 2001, 79:138–155.
[27]Sun T, Li Y, Xiao Z, et al. Semiparametric Copula Method for Semi-competing Risks Data Subject to Interval Censoring and Left Truncation:Application to Disability in Elderly[J]. Statistical Methods in Medical Research, 2023, 32(4):656–670.
[28]Zheng M, Klein J P. Estimates of Marginal Survival for Dependent Competing Risks Based on An Assumed Copula[J]. Biometrika, 1995, 82:127–138.
[29]Zhou Q, Hu T, Sun J. A Sieve Semiparametric Maximum Likelihood Approach for Regression Analysis of Bivariate Interval-Censored Failure Time Data[J]. Journal of the American Statistical Association, 2017, 112(518):664–672.
(1)数据来源为国家统计局网站，网址链接为https://www.stats.gov.cn/sj/zxfb/202502/t20250228_1958817.html。
(2)网址链接为https://population.un.org/wpp/publications。
(3)网址链接为https://www.ndrc.gov.cn/fggz/fzzlgh/gjjzxgh/202203/t20220325_1320220.html。
(1)常见的阿基米德Copula函数主要有Frank Copula、Clayton Copula、Gumbel Copula。
(1)这些样本不符合次序多状态模型假设，故而予以删除。
(2)因篇幅所限，样本中不同健康状态转移路径的个体数量及占比以附表1展示，见《统计研究》网站所列附件。下同。
(3)因篇幅所限，样本数据删失情况以附表2展示。
(1)因篇幅所限，分性别75岁轻度失能老年人的后序事件累积发生概率以附图1展示。
(1)因篇幅所限，老年人随年龄变化的两年状态转移概率堆积图以附图2展示，男性老年人随时间变化的状态转移概率堆积图以附图3展示。
基本信息:
DOI：10.19343/j.cnki.11-1302/c.2026.02.011
中图分类号:R195;F842.6
引用信息:
[1]夏贺彦,亢妍,王晓军.基于相依删失数据的中国老年人健康状态转移概率估计[J].统计研究,2026,43(02):144-156.DOI:10.19343/j.cnki.11-1302/c.2026.02.011.
基金信息:
教育部哲学社会科学研究重大课题攻关项目“健康中国2030背景下的健康老龄化体系优化研究”（20JZD023）; 国家自然科学基金面上项目“面向老年失能风险管理的复杂多状态生存数据建模及应用”（72471229）
2026-02-25
2026-02-25